This invention relates generally to the synthesis of alcohols and ethers from alkanes, and more particularly to a method of and apparatus for manufacturing methanol and dimethyl ether from methane.
Methane has previously been converted to methanol by the halogenation of methane followed by hydrolysis of the methyl halide to form methanol. For. example, gaseous chlorine has been used to chlorinate methane to form chlorinated methane, principally methyl chloride, together with other chlorides, i.e., dichloromethane, trichloromethane and carbon tetrachloride. Alternatively, methane has been subjected to oxychlorination with oxygen and hydrochloric acid to form the foregoing compounds. The chlorinated methanes produced are hydrolyzed in the vapor phase to produce methanol, formaldehyde, formic acid and by-products, including carbon dioxide and hydrochloric acid, depending on the chlorination selectivity. Hydrochloric acid is produced or used in the halogenation of methane by either method and must be recovered, dehydrated by azeotropic distillation and recycled. Corrosion and other problems involved with the handling of chlorine and hydrochloric acid are substantial.
U.S. Pat. No. 3,172,915 granted to Borkowski, et al. is directed to a process for converting methane to methanol. Borkowski discloses the chlorination of methane using ferric chloride at high temperatures to produce chloromethanes and hydrogen chloride. The process requires temperatures in the range of 220-8002xc2x0 C., more preferably 250-450xc2x0 C., and long residence times, e.g., more than one hour. Further, the process is hindered by the production of a mixture of chlorination products, e.g., chloromethane, dichloromethane, trichloromethane and carbon tetrachloride, which must be separated before hydrolysis to methanol. Other disadvantages result from the energy required to dry the ferric chloride and from the corrosion and handling problems inherent with hydrochloric acid.
U.S. Pat. No. 5,243,098 granted to Miller discloses another method for converting methane to methanol. In the Miller process the reaction of methane with cupric chloride produces chloromethane and hydrochloric acid. These intermediates are then reacted with steam and a catalyst containing magnesium oxide to produce methanol and magnesium chloride. Magnesium oxide is regenerated by treatment of the magnesium chloride by-product with air or oxygen. Cupric chloride is regenerated by treatment of the cuprous chloride by-product with air and hydrochloric acid. While these reactions proceed at favorable rates, attrition of the solid reactants, i.e., cupric and magnesium oxide, is significant. Special filters and processes are required to recover and regenerate the reactants in the required particle size. Miller also suggests cupric bromide and magnesium zeolite as alternative reactants. Because of the attrition of the reactants, difficulties associated with the handling of solids, and the special filters and processes required to regenerate the reactants, the Miller process has proven unsatisfactory. U.S. Pat. No. 5,334,777, also granted to Miller, discloses a nearly identical process for converting ethane to ethylene glycol.
U.S. Pat. No. 5,998,679 granted to Jorge Miller, discloses a process for converting alkanes and alkenes to the corresponding lower alkanols and diols. In the method of the invention, a gaseous halogen (bromine) is produced by decomposing a metal halide in a liquid having a melting point below and a boiling point above the decomposition temperature of the metal halide. The preferred liquid is molten hydrated ferric chloride maintained at a temperature between about 37-280xc2x0 C. The lower alkane or alkene is halogenated in a gas phase reaction with the halogen. The resulting alkyl halide or alkyl dihalide is contacted with a metal hydroxide, preferably an aqueous solution of ferric hydroxide, to regenerate the metal halide and produce the corresponding lower alkanol or diol. Problems with this process include low monohalogenation selectivity, and corrosiveness of the hydrated ferric halides, which may present a containment problem if the process is run at 280xc2x0 C., where high pressures of steam are required to maintain ferric halide hydration. Finally, the process produces a great deal of water and HCl or HBr, all of which are difficult to separate on a large scale from the desired product methanol.
Published international patent application WO 00/07718, naming Giuseppe Bellussi, Carlo Perego, and Laura Zanibelli as inventors, discloses a method for directly converting methane and oxygen to methanol over a metal halide/metal oxide catalyst. This is not a catalyst in the true sense of the word, however, because the reaction involves transfer of halide from a metal halide via reaction with methane to a different metal oxide producing the metal halide and methanol downstream. Eventually the halide is leached and the catalyst loses activity.
Olah et al. (George A. Olah, et al. J. Am. Chem.
Soc. 1985, 107, 7097-7105) discloses a method for converting methane to methanol via methyl halides (CH3Br and CH3Cl), which are then hydrolyzed to prepare methanol. In the process, CH3Br and CH3Cl are hydrolyzed over catalysts with excess steam generating a methanol, water, and HCl or HBr mixture. The separation of methanol (about 2% by mole) from HCl or HBr and water on an industry scale (2000 tons per day) requires an enormous amount of energy and generates a great deal of aqueous HCl or HBr waste. Aqueous HCl and HBr are very corrosive as well.
The present invention comprises a process wherein bromine or a bromine containing compound is used as an intermediate to convert alkanes to alcohols, ethers, or olefins by reaction with oxygen or air. While the process can be used to convert a variety of alkanes, including methane, ethane, propane, butane, isobutane, pentanes, hexanes, cyclohexane, etc. to their respective alcohols, ethers, or olefins, the conversion of methane to methanol and dimethyl ether is illustrative.
Methane reacts with bromine over a catalyst to form CH3Br and HBr. CH3Br and HBr react with a metal oxide to form a variable mixture of dimethyl ether (DME), water and methanol, and the metal bromide. The metal oxide and molecular bromine are regenerated by reaction of the metal bromide with air and/or oxygen. The regenerated bromine is recycled to react with methane while the regenerated metal oxide is used to convert more methyl bromide to methanol and DME, completing the reaction cycle.
The process can be easily carried out in a riser reactor. Compared to the current industrial two step process, in which methane and steam are first converted to CO and H2 at 800xc2x0 C. followed by conversion to methanol over a Znxe2x80x94Cuxe2x80x94Alxe2x80x94O catalyst at approximately 70-150 atmospheres, the process of the present invention operates at roughly atmospheric pressure and relatively low temperatures, thereby providing a safe and efficient process for methanol production.
The present invention operates with solid/gas mixtures at atmospheric pressure. In the process, the hydrogen halide is gaseous, and therefore not as corrosive as when aqueous at high temperatures. The reaction of Br2 with an alkane can reach more than 90% selectivity and high conversion to alkane-monobromide. The main side products, alkane dibromides such as CH2Br2 can be converted back to the monobromides by reaction with an alkane over another catalyst. Very few by-products are produced.
In the process most of the Br atoms are trapped in the solid state, making the system less corrosive.
Another advantage is that DME and alcohol (CH3OH) are not produced as a mixture with excess water. By controlling reaction conditions, almost pure DME and/or methanol is obtained directly so that it is not necessary to separate CH3OH from water. Finally, in the present process, methane and oxygen do not come into direct contact, resulting in improved safety.